Zoo Biology 15:509-524 (1996)

Nutrition of the Tamandua: I. Nutrient

Composition of Termites (Nasutitermes
spp.) and Stomach Contents From Wild
Tamanduas (Tamandua tetradactyla)
Sergio E. Oyarzun, Graham J. Crawshaw, and Eduardo V. Valdes

Metropolitan Toronto Zoo, Scarborough, Ontario, Canada

Arboreal termites (Nasutitermes spp.) and stomach contents from tamanduas (Tamandua tetradactyla)
were collected in central Venezuela during the mid part of the dry season (March) of 1993 and 1994.
Nutritional analyses were performed on each caste (workers (n = 3), soldiers (n = 5), and alates (n =
1]), on mixed caste samples (n = 1), and on stomach contents from live (n = 5) and roadkill (n = 5)
tamanduas. The chemical analysis, expressed on a dry matter (DM) basis, of termite workers, which
constituted the majority of the nest populations, showed the highest crude protein (CP) (67%) and the
lowest DM (25%) and fat (2%) values. Ash content varied from a low of 4% in alates to a high of 7%
in soldiers. The alates contained substantially higher DM (41%) and fat (40%), which was reflected in
a higher caloric value (6.88 kcal/g) (gross energy) [GE]), and relatively less CP (49%). Among the
macrominerals, potassium (K) was consistently the highest, with an overall mean value of 0.54%,
while the calcium (Ca) and phosphorus (P) levels showed overall means of 0.26% and 0.67%,
respectively. Iron (Fe) was the highest among the trace minerals but highly variable (soldiers, 1,000
ppm; alates, 246 ppm; workers, 394 ppm). Differences in the concentrations of vitamin A and E were
found among termites castes, with soldiers showing the highest values (20 and 85 µg/g for retinol and
a-tocopherol, respectively). Acid detergent fiber (ADF) was lower in the alates (13%) and workers
(27%) compared to the soldiers (35%). Alates’ fat was more saturated (39%), while soldiers and
workers had a much higher polyunsaturated fatty acid (PUFA) concentration. In general, similar
nutrient profiles were found between the tamandua stomach contents and the overall mean
composition of Nasutitermes spp. However, stomach contents had much higher ADF, ash, and Fe
concentrations (3 1%, 14%, and 2,748 ppm) than termites (25%, 5%, and 652 ppm) but lower CP, fat,
GE, and Ca values (51%, 11%, 4.58 kcal/g, and 0.11% vs. 58%, 15%, 6.01 kcal/g, and 0.26% in
termites). The relatively low concentrations of Ca in both stomach contents and termites may be
indicative of a low requirement in Myrmecophaga compared to other mammalian species. Diets
consumed by free-ranging

Received for publication June 30, 1995; final revision accepted June 17, 1996.
address reprint requests to Sergio E. Oyarzun, 1704 Wollaston Court, Pickering, Ontario, Canada
L1V 2X1.
Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
510 Oyarzun et aI.

tamanduas contained on a DM basis 51% CP, 11% fat, 14% ash, 31% ADF, 4.58 kcal/g GE, 0.11%
Ca, 0.41% P, 2.52 µg/g retinol, and 44.3 µg/g α-tocopherol. Duplication of these nutrient profiles
might greatly benefit captive health and reproduction of this species.

Key words: Myrmecophaga, anteater, natural diet, nutrients, arboreal termites

INTRODUCTION

Feeding insectivorous mammals in captivity has always presented a significant challenge. Both the
nutritional and behavioral requirements must be met. Very little is known about the nutritional needs of these
species, and their nutrient requirements and tolerance levels have not been established. Although a few
studies have been carried out on the dietary habits of anteaters [Montgomery, 1985a,b; Redford, 1985],
information on the chemical composition of the prey foods has been restricted to gross components
(moisture, protein, fat, and ash) [Phelps et al., 1975; Matsumoto, 1976; Redford and Dorea, 1984], and no
information is available on the mineral, vitamin, amino acid, and fatty acid composition of ants and termites.
Anteaters of the genus Tamandua are highly specialized predators, subsisting exclusively on a diet
of ants and termites [Lubin and Montgomery, 1981; Montgomery, 1985b]. There are two species: Tamandua
mexicana, which originates from southern Mexico to northwestern South America, and T. tetradactyla,
which is distributed from northwestern Venezuela to northern Argentina and Uruguay [Wetzel, 1975].
Tamanduas hunt primarily by scent, moving almost continuously during an active period lasting 8 -
10 hr, searching for prey and stopping to feed infrequently for short periods of time (<1 mm). They are
individualistic in the timing of their activity, in their use of different microhabitats and degree of arboreality,
and in the species composition of their diet [Lubin et al., 1977; Montgomery and Lubin, 1977; Lubin and
Montgomery, 1981; Redford, 1985]. They consume both ants and termites, with an apparent preference for
the reproductive and worker castes over the soldier caste [Lubin and Montgomery, 1981].
North American and South American zoos have a poor record at keeping tamanduas. Small
numbers have been maintained in captivity for several years, but reproduction is uncommon. Poor survival
during earlier years is likely related to their specialized dietary requirements and digestive problems [Meritt,
1975, 1976]. The Metropolitan Toronto Zoo (MTZ) has been successful at maintaining the species for
extended periods of time, although no reproduction has occurred and a significant medical problem has been
observed in our animals. Three T. mexicana arrived at the MTZ in 1981, and one of them is still alive. In
1987, a male showed signs of rear limb paresis and urinary retention, progressing to complete flaccid
paralysis. Radiography revealed extensive hyperostosis (excessive growth or thickening of bone tissue) of
the thoracic, lumbar, and coccygeal vertebrae.
Radiographs of the other animals revealed similar bone lesions. Two wild caught T. tetradactyla
were added to the MTZ collection in 1986. Both of these younger animals, normal when first examined,
subsequently developed hyperostosis of the axial skeleton. Similar lesions have been observed in tamanduas
in European zoos [Dierenfeld et al., 1995] and in tamanduas and the giant anteater in other North American
zoological institutions.

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
 Nutrition of the Tamandua 511

The pathological lesions were suggestive of either hypervitaminosis A or D or perhaps excessive

calcium (Ca) intake, but interpretation of findings from affected animals was difficult due to the lack of
normal biochemical and nutritional data [Crawshaw and Oyarzun, in press].
A recent survey on diets fed to tamanduas and giant anteaters in Brazilian and North American zoos
revealed a wide variation in their calculated analysis. On a dry matter (DM) basis, Ca levels ranged from
0.04-2.34%, vitamin A from 0.5-57.3 IU/g, and vitamin D from 0.05-68.8 IU/g [Ann Ward, personal
communication].
At MTZ the tamanduas were initially fed a diet similar to that recommended by Meritt [1970],
consisting of a mixture of evaporated milk, canned dog food, Gevral protein, and a vitamin-mineral
supplement. Subsequently, after presentation of the symptoms, various diet adjustments were introduced, and
between 1987 and 1994 they were fed the zoo’s carnivore mixture (300 g/animal/day).
This investigation forms part of a series of studies on the physiology and nutrition of tamanduas
undertaken by the Animal Nutrition Centre of the MTZ to assess the nutritional status of free-ranging
tamanduas. The main objective of the present study was to gather baseline information on the nutrient
composition of the natural diet of the tamandua through the collection and chemical analysis of one of its
primary foods (termites) and of stomach contents from live and roadkill specimens. New diet
recommendations may then be made based on this information, thereby improving survival and reproduction
of this species in captivity.

MATERIALS AND METHODS

Capture of Tamanduas
The study was conducted in the central llanos (savannahs) of Venezuela during the mid part of the
dry season (March) in 1993 and 1994. The search for live tamanduas was carried out mainly at night from
2100 to 0500 hr in three different habitats according to the vegetation communities. These were a gallery
forest and a sparsely vegetated palm savannah on Hato Masaguaral (a cattle ranch located 45 km south of
Calabozo, Guárico State) and the savannahs and forests of Hato Piñero, a cattle ranch located near the town
of El Baál (Cojedes State).
Some animals were caught by their tails, lifting them up from the ground or from lower tree
branches, while others had to be darted to bring them down from the trees. The animals were anesthetized
with ketamine (Ketaset; Ayerst, St. Laurent, Quebec, Canada), 11 mg/kg, and xylazine (Rompun, Bayvet,
Etobicoke, Ontario, Canada), 0.8 mg/kg IM. Immobilized animals were examined, sexed, weighed, and
measured. After samples of blood, hair, and stomach contents were taken and upon recovery from anesthesia,
the animals were released at the same capture locations.

Stomach Contents
Stomach contents were collected from five live adult tamanduas captured on Hato Masaguaral. An
82 cm x 6 mm diameter rubber feeding tube (CDMV, Montreal, Quebec, Canada) was passed into the
stomach. The contents were flushed with water and aspirated into a 60 ml catheter-tipped syringe (Becton-
Dickinson, Mississauga, Ontario, Canada) and transported to the ranch facilities for processing. Excess water
was removed, and the contents were stored in cryogenic vials and frozen in liquid nitrogen.

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
 512 Oyarzun et al.

Thirteen carcasses of roadkill tamanduas were obtained along the roads of Guárico, Cojedes,
Portuguesa and Lara states, during the 1993 and 1994 study periods. The quality of the carcasses ranged
from very fresh to moderately autolyzed, but all had been killed within a 12 hr period. Stomach contents
were collected only from five fresh carcasses. The stomachs were removed from the carcasses and dissected
and the entire contents emptied into plastic containers, temporarily kept on ice, and subsequently transferred
into cryogenic vials. All samples were stored in liquid nitrogen until analysis. Stomach contents from two
animals were examined for identification of insect remains. All tamanduas were T. tetradactyla except for
one road-kill, T. mexicana, collected in the area of Maracaibo (Zulia State).

Termite Collection
During the 1993 field study, samples of arboreal termites were taken from two colonies, one on
Hato Masaguaral, the other near the town of Acarigua, Portuguesa state. Collection was achieved by slicing a
section of the carton nests with a machete, causing the termites to come out and defend the colony. They
were then brushed off into temporary plastic containers and subsequently packaged in cryogenic vials and
frozen in liquid nitrogen. These two samples were comprised exclusively of nasute soldiers and classified as
Nasutitermes corniger [Maribel Hernandez, personal communication]. The social behavior of the termites
forced us to review our collection methods, as all castes of the termite colony were not represented in our
samples.
When a termite nest is disturbed, the nasute soldiers concentrate in great numbers at the site of the
break, and the workers disappear into the interior of the nest or the covered trails [Lubin and Montgomery,
1981]. Thus, a different approach was followed in 1994. Entire colonies were collected by removing whole
arboreal carton nests (two on Hato Piñero and one on Hato Masaguaral) and transporting them in plastic bags
to the ranch facilities. They were then fractioned into small sections from which the termites were shaken or
picked with forceps, stored in plastic bags, and cleaned from any debris. Using the distinct morphological
characteristics of the members in a termite colony, they were sorted into the different castes (soldiers,
workers, alates), separately packaged in cryogenic vials, and preserved in liquid nitrogen until analysis. Only
one nest contained mature alates (nest 1 from Hato Piñero). The other two colonies contained predominantly
the soldier and worker castes. All termites were alive at the time of freezing.

Termite Nest Material

Nest materials were collected from one of the nests taken on Hato Piñero and one nest from
HatoMasaguaral. The nests were broken up into small pieces and packaged in air-tight plastic bags for future
analysis.

Chemical Analyses
All termite and stomach content samples were lyophilized (Labconco, Freeze Dryer 18; Labconco
Corp., Kansas City, MO). The freeze-dried termite and stomach content samples, as well as the nest material
samples, were ground through a 1 mm mesh screen (Tecator Cyclotec Sample Mill; Fisher Scientific,
Montreal, Quebec, Canada).
Samples were analyzed for DM, fat, ash, and crude protein (CP) according to the methods of the
AOAC [1990]. Gross energy (GE) was determined in an adiabatic bomb calorimeter (Parr Instrument

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
 Nutrition of the Tamandua 513

Company, Moline, IL). Absolute DM determination was done by drying to a constant weight in a vacuum
oven at 1000C. The same samples were placed in cellulose thimbles (Whatman, W & R Balston Ltd.,
London, England) for fat extraction using anhydrous ethyl ether in a Soxhlet extractor.
The dry and fat-free samples were wet-ashed with a mixture of concentrated nitric and perchloric
acid. Mineral analysis (Ca, Mg, K, Na, Fe, Zn, Cu, and Mn) was performed by atomic absorption
spectrophotometry (Perkin-Elmer model 2380; Perkin-Elmer Corp., Norwalk, CT) [Emerson, 1975].
Phosphorus (P) was determined by the alkalimeter animonium molybdate method [AOAC, 1980]; and
selenium (Se) was determined by the Fluorimetric Method [Hoffman et al., 1968] using a Perkin-Elmer
model 3000 FluorescenceSpectrometer. Validation of the mineral data was acc9mplished using a standard
reference material (bovine liver #1 577a; National Bureau’ of Standards, Gaithersburg, MD).
Nitrogen (N) determination was performed using a nitrogen analyzer (Leco Instruments Ltd.,
Mississauga, Ontario, Canada), and the total N by a factor of 6.25 gave the total CP values. Fatty acid
analysis was done in a Hewlett-Packard 5890 Series II gas chromatograph (Hewlett-Packard Co., Avondale,
PA). Fatty acid methyl esters were identified by comparison of retention times with standards (Nu-
CheckPrep, Elysian, MN) and expressed as percentages of total methyl esters.
Termite and tamandua stomach content samples were hydrolyzed for 24 h with 6 N HC1 at 1100C
for the determination of amino acid (AA) levels using the Pico-Tag system (Waters Chromatogrpahy
Division, Millipore Corporation, Milford, MA) as described in Sarwar et al. [1988]. The sulphur-containing
AA are underestimated because samples were not preoxidized before hydrolysis. The fiber components were
determined by using the general methods of Van Soest [1982, 1991].
Retinol and α-tocopherol were analyzed by high performance liquid chromatography (HPLC) using
a Varian Vista 5500 Liquid Chromatograph equipped with Vista 402 microprocessor and model 8058
autosampler (Varian Canada Inc., Montreal, Quebec, Canada) as described for feed samples in sections 43
.008 - 43 .013 and 43.064 - 43.068 [AOAC, 1975].
All analyses were performed in duplicate at the Crampton Nutrition Laboratory of McGill
University (Montreal, Quebec, Canada), and all calculations were conducted on an absolute DM basis.

RESULTS AND DISCUSSION

Proximate Analysis, Gross Energy, Fiber, and Vitamin Content In Termites
The proximate analysis, GE, fiber fraction, and vitamin contents of the different termite castes
(mean ± SEM) expressed on DM basis are presented in Table 1.
The reproductive caste (mature alates) had a considerably higher DM (41.0%) and fat content
(40.2%) and a lower ash concentration (3.7%) than the other termite castes, which was also reflected in a
higher caloric value (6.88 kcal]g). The fat content of the alates was considerably higher than the fat values
reported by Redford and Dorea [1984] for immature alates (nymphs) of four other termite species (range:
19.7 - 24.1%) but quite similar to the fat level (v.3%) found in mature (wings removed) alates of
Macrotermes falciger [Phelps et al., 1975]. CP content of alates (48.8%) was substantially lower than the

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
514 Oyarzun et al.

TABLE 1. Proximate analysis, gross energy, fiber and vitamin concentrations in termite (Nasutitermes
spp.) soldiers, workers, alates, and mixed castes*
Soldiers Workers Alates Mixeda
Analysis (n = 5) (n = 3) (n = 1) (n =1) Overall
Dry matter (%) 30.43 ±2.58 24.70 ± 0.51 41.00 21.29 29.36 ± 4.32
Gross energy (kcal/g) 5.29b — 6.88 5.87 6.01 ±0.46
Crude protein (%) 58.03 ±6.46 66.71 ± 2.70 48.80 59.28 58.20 ± 3.67
Crude fat (%) 11.23 ±4.27 2.21 ± 1.05 40.23 6.50 15.04 ± 8.6
Ash (%) 3.72c ±0.34 4.58c ± 0.22 3.72 4.42 4.11 ± 0.23
NDF (%) 37.55 ±0.43 — 23.37 30.77 30.56 ± 4.09
ADF (%) 34.81 ±1.08 27.09 ± 0.32 13.02 25.44 25.09 ± 4.51
Cellulose (%) 11.31 ±1.41 — 6.35 11.64 9.77 ± 1.71
Lignin (%) 23.51 ±2.49 — 13.03 15.22 17.25 ± 3.19
ADF-N (%) 1.20b 3.54 ± 0.04 3.99 3.91 3.16 ± 0.66
Retinol (µg/g) 20.4c ±6.14 n.d.d 0.65 1.20 7.42 ± 6.49
α-tocopherol (µg/g) 84.45c ±15.71 n.d.d 40.44 152.60 92.50 ±32.63
*Means ± standard errors. All values expressed on a dry matter basis. A dash indicates there was not sufficient sample.
a
Mixed castes; sample contained approximately 90% workers and 10% soldiers.
b
n = 1.
c
n = 2.
d
nd = not detectable amounts.

values for soldiers, workers, and a mixed sample (58.0, 66.7, and 59.3%, respectively) but consistent with
those reported for alates of other termite species [Phelps et al., 1975; Matsumoto, 1976].
Termite soldiers had similar DM, CP, Ca, P, and GE value to the larvae of the mulberry silk moth
[Frye and Calvert, 1989]. The high CP values as listed in Table 1 may be misleading, since they are derived
from measurements of all the N, including that trapped in the chitinous exoskeletons which most probably is
not available to the tamandua.
In general, the gross composition (DM, ash, CP, and GE) of the Nasutitermes spp. alate caste, as
determined in this study (Table 1), is in reasonable agreement with those previously reported by Matsumoto
[1976] for alate nymphs of three species of epigeous nest builder termites from West Malaysia (average
44.0% DM, 1.7% ash, 38.0% CP, and 6.8 kcal/g) and the CP (41.8%) and GE (7.6 kcal/g ash-free DM)
values of M. falciger alates as reported by Phelps et al. [1975].
The workers had the highest CP content (66.7%), followed by the soldiers with a mean CP value of
58.0%. The analysis of the mixed sample (59.3% CP), which contained approximately 90% workers, tends to
confirm these results. However, these findings are in disagreement with previously reported values for the
same genus of termites. Redford and Dorea [1984] found that Nasutitermes workers had significantly lower
CP content than soldiers (44.6 vs. 52.4%). The same authors also reported that, with the exception of
Cortaritermes silvestri, the worker caste of all species studied had consistently lower CP levels than soldiers
of the same species. Similar findings have been reported by Matsumoto [1976] for epigeal termites from
West Malaysia (69.4 vs. 55.8% on an ash-free DM basis for soldiers and workers, respectively). These
differences are unexplainable and warrant further investigation.
A very large variability was found in the vitamin content of the different castes (Table 1). The
concentration of retinol and a-tocopherol in soldiers was considerablyhigher (20.4 and 84.5 µg/g,

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
 Nutrition of the Tamandua 515

respectively) than in the alates (0.7 and 40.4 µg/g), contrary to what may be expected considering the much
higher fat concentration in the alates. The reported vitamin values for soldiers were obtained from samples
collected during the 1993 field study. It is interesting to note that the analysis performed on the three samples
of workers showed no detectable amounts of these vitamins, while the mixed sample (approximately 90%
workers) had a retinol value 1.2 µg/g and the highest a-tocopherol value (152.6 µg/g). Considering the small
number of samples analyzed, these results should be considered inconclusive and warrant further
investigation.
The overall mean retinol (7.42 µg/g) and a-tocopherol (92.5 µg/g) levels rep resent 24,733 IU/kg
and 137.8 IU/kg of vitamin A and vitamin E activity, respectively (conversion factors: 0.3 µg retinol = 1 IU;
1 mg α-tocopherol = 1.49 IU). These values exceed the requirements of dogs (5,000 IU and 50 IU/kg DM)
and cats (10,000 IU and 80 lU/kg DM) for these vitamins [National Research Council, 1985, 1986] and most
likely the needs of the tamandua.
Assays for ascorbic acid were also performed, but no detectable amounts were found in any of the
termite or tamandua stomach contents samples.
The assays for fiber [Van Soest, 1982, 1991] have been developed specifically for determination of
the fiber fractions in plant materials. Their application for analyzing insect matter is questionable. These
fiber fraction values perhaps reflect the complex carbohydrate content of the termite exoskeletons (including
chitin) that mimic plant components, but it may also represent a measure of true fiber values of the termites’
digestive contents, as these insects feed on wood, a cellulose- and lignin-rich material. The acid detergent
fiber (ADF) was lower in the softer bodied alate (13.0%) and worker (27.1%) castes compared to the harder,
more schlerotized soldiers (34.8%), consistent with what may be expected (Table 1). The concentration of
“lignin” in soldiers was almost double (23.5%) the values of alates and mixed termites (13.0 and 15.2%,
respectively).

Mineral Concentrations In Termites

The ash content and the mineral profiles (mean ± SEM) of the different castes of Nasutitermes spp.
termites, expressed on a DM basis, are shown in Table 2.
Analysis of ash content indicated that the worker caste had a higher ash value (4.6%) than soldiers
and alates (3.7%), while the mixed caste sample had a concentration (4.4%) closer to the worker samples
(Table 2). The higher ash content of the worker caste is consistent with previous reports [Matsumoto, 1976;
Redford and Dorea, 1984]. Redford and Dorea [1984] stated that the probable reason for the differences
between castes lies in the feeding biology of termites; since the workers ingest the food and feed the soldiers,
the former would be expected to have higher concentrations of indigestible materials and thus a higher ash
content.
The absolute ash values of workers and soldiers listed in Table 2 are in disagreement with those
previously reported by Redford and Dorea [1984]. These authors reported a considerably higher ash
concentration in Nasutitermes spp. termite workers and soldiers (11.3 and 8.8%, respectively) and even
higher values in six other species of carton nest-builder termites (22.7 and 11.3% for workers and soldiers,
respectively), while the mean ash content in two species of geophagous termites was 60.5% for workers and
18.4% for soldiers.
A report on chemical composition of epigeous nest-builder termites by Matsumoto [1976] also

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
516 Oyarzun et al.

TABLE 2. Mineral content of Nasutitermes spp. termites*

Soldiers Workers Alates Mixeda
Analysis (n=2) (n=2) (n=1) (n=l) Overall
Ash, % 3.72b ± 0.34 4.58b ± 0.22 3.72 4.42 4.11 ± 0.23
Macro minerals (%)
Calcium 0.37 ± 0.1 0.20 ± 0.03 0.24 0.22 0.26 ± 0.04
Phosphorus 0.29 ± 0.04 0.40 ± 0.04 0.36 0.46 0.38 ± 0.04
Magnesium 0.15 ± 0.03 0.13 ± 0.03 0.15 0.13 0.14 ± 0.01
Potassium 0.58 ± 0.02 0.61 ± 0.02 0.37 0.60 0.54 ± 0.06
Sodium 0.06 ± 0.01 0.24 ± 0.02 0.21 0.17 0.17 ± 0.04
Trace minerals (ppm)
Iron 1,001 ± 736 394 ± 12 246 965 652 ± 194
Zinc 164 ± 16 144 ± 2 184 159 163 ± 8
Manganese 115 ± 27 32 ± 4 37 46 57 ± 20
Copper 33 ± 9 52 ± 14 18 50 38 ± 8
Selenium 0.51 ± 0.18 — — — 0.51 ± 0.18
*Means ± standard errors. All values expressed on a DM basis. A dash indicates there was not sufficient sample.
a
Mixed castes; sample contained approximately 90% workers and 10% soldiers.
b
n = 3.

indicated much higher ash concentrations in workers (26.0 - 66.0%) than in soldiers (2.0 - 24.0%). The
differences were attributed to the fact that epigeal termite workers transport in their alimentary system vast
amounts of mineral and organic components of soil to build and maintain their mounds.
The mineral profiles appear to be more consistent among the different termite castes. Potassium (K)
was the highest by far among the macro minerals, with Ca values less than half the amount of K, distinctive
from vertebrate species with higher Ca and P content. But this is logical, as invertebrates have no bony
skeletons. Soldier, worker, and mixed caste K values were almost identical (0.58, 0.61, and 0.60%,
respectively), while alates had a lower K concentration (0.37%). The soldier caste had the highest Ca
(0.37%) and lowest P concentration (0.29%), providing an adequate Ca:P ratio of 1.28, while the Ca values
for the other castes and mixed sample were determined a little over 0.2%, with consistently higher P levels
(0.36 - 0.46%), giving poor Ca:P ratios of 0.5 - 0.7. These results indicate that termites alone may be
marginally adequate in providing the Ca needs of insectivorous mammals in general and the tamandua in
particular.
Iron (Fe) values were the highest among the trace minerals but highly variable. The concentration of
this element in the soldier caste (1,001 ppm) was two times higher than Fe content in workers (394 ppm) and
exceeded by fourfold the Fe content in the alates (246 ppm) (Table 2).
Manganese (Mn) concentration in soldiers (115 ppm) exceeded by twofold the levels found in the
other termite castes. The concentrations of zinc (Zn) and copper (Cu) were less variable among castes,
ranging from 144 - 184 ppm and 18 - 52 ppm, respectively. Se levels were determined in only the two
samples of soldiers collected in the 1993 field study and averaged 0.51 ± 0.18 ppm (Table 2).

Fatty Acid Profile of Termites

The fatty acid profiles of the lipid fraction in the termite samples indicate that the alates’ fat was
more saturated (38.6%) than the fat in soldiers (33.5%) and workers (31 .8%),

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
 Nutrition of the Tamandua 517

TABLE 3. Analyzed fatty acid composition of the lipid fraction in Nasutitermes spp. termites
(mean ± SEM)
Soldiers Workers Alates Mixed
Fatty Acida (%) (n = 2) (n = 2) (n = 1) (n = 1) Overall
Fat soxhlet (%) 11.23 ± 4.27 2.21 ± 1.05 40.23 6.50 15.04 ± 8.6
Capric 10:0 1.24 ± 0.4 0.53 ± 0.22 0.04 1.05 0.55 ± 0.5
Lauric 12:0 0.43 ± 0.3 1.93 ± 1.78 0.16 0.06 0.11 ± 0.05
Myristic 14:0 4.27 ± 1.64 4.10 ± 0.16 4.89 3.46 4.18 ± 0.71
Myristoleic 14:1(n-7) 1.68 ± 0.58 2.09 ± 0.36 0.19 1.64 0.92 ± 0.73
Palmitic 16:0 7.54 ± 1.23 9.71 ± 0.38 19.15 7.78 13.47 ± 5.69
Palmitoleic l6:l (n-7) 0.99 ± 0.2 2.11 ± 1.34 1.42 0.64 1.03 ± 0.39
Stearic 18:0 11.81 ± 1.4 12.51 ± 0.67 12.68 12.43 12.56 ± 0.12
Oleic l8:1 (n-9) 32.63 ± 2.9 48.86 ± 1.81 51.08 44.64 47.86 ± 3.22
Linoleic l8:2 (n-6) 11.08 ± 1.5 15.05 ± 1.33 8.51 15.29 11.90 ± 3.39
Linolenic 18:3 (n-3) 9.67 ± 0 0.00 0.00 5.17 2.59 ± 2.59
Arachidic 20:0 3.87 ± 0.17 1.66 ± 0.29 1.30 2.02 1.66 ± 0.36
Gadoleic 20:1 (n-11) 1.26 ± 0 0.00 0.20 0.00 0.10 ± 0.1
Arachidonic 20:4 (n-6) 9.20 ± 0.93 0.00 0.05 3.48 1.77 ± 1.72
Behenic 22:0 4.38 ± 4.09 1.40 ± 0.25 0.33 1.52 0.93 ± 0.6
SATb 33.52 ± 3.97 31.82 ± 2.08 38.55 28.32 33.44 ± 5.12
MONOb 36.55 ± 3.41 53.06 ± 0.83 52.89 46.92 49.91 ± 2.99
PUFAb 29.94 ± 0.56 15.05 ± 1.33 8.56 23.94 16.25 ± 7.69
a
Fatty acid content expressed as percentages of total methyl esters. Fatty acids are denoted by their common name,
number of carbons:number of double bonds, followed by the position of the first double bond relative to the methyl-
end (n-).
b
SAT = sum total percentage of 10:0, 12:0, 14:0, 16:0, 18:0, 20:0, and 22:0; MONO = sum total percentage of 14:1,
16:1, 18:1, and 20:1; PUFA = sum total percentage of 18:2, 18:3, and 20:4.

while soldiers and workers had much higher concentration of polyunsaturated fatty acids (PUFA) (29.9 and
15.1%, respectively) than the alates (8.6%) (Table 3).
Oleic acid was the predominant fatty acid in the lipid fraction, irrespective of the termite class, with
an overall average of 47.9 ± 3.2%. The soldiers’ fat had the highest concentration of arachidonic acid (9.2%),
followed by the mixed caste sample (3.5%), while alates’ fat had less than 0.1%. Workers’ fat had no
measurable amounts of arachidonic acid. However, these results are probably not conclusive, considering the
small number of observations.

Amino Acid Profile of Termite Soldiers and Tamandua Stomach Contents

The AA profiles (mean ± SEM) of tamandua stomach contents (n = 4) and termite soldiers (n = 2)
are summarized in Table 4. AA assays of termite workers and alates were not done, as no sample was
available. The AA profile of Macrotermes falciger alates [adapted from Phelps et al., 1975] is included for
comparative purposes.
Sample analysis indicated considerable differences in CP content on a DM basis (50.9 vs. 43.7%)
between the taniandua stomach contents and the termite soldiers collected in the 1993 field study (Table 4).
The tamandua stomach contents also had a much higher CP value than the alates of M. falciger (41.8%), as
reported by Phelps et al. [1975]. The CP values for the nasute soldiers (43.7%) are much lower than those
previously reported for the same kind of termite (5 2.0%) by Redford and Dorea

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
518 Oyarzun et al.

TABLE 4. Amino acid profile of stomach contents from Tamandua tetradactyla as compared to the
amino acid content of Nasutitermes spp. soldiers and Macrotermes falciger alates (mean ± SEM; all
values expressed as a percentage of dry matter)
Stomach contents Termite soldiersa M. falciger alatesb
(n=4) (n=2) (n=2)
Essential (EAA) (%)
Arginine 2.47 ± 0.11 1.44 ± 0.03 2.57 ± 0.15
Histidine 1.11 ± 0.08 0.69 ± 0.08 1.28 ± 0.03
Isoleucine 1.65 ± 0.08 0.77 ± 0.05 1.69 ± 0.02
Leucine 3.07 ± 0.14 1.57 ± 0.08 3.12 ± 0.02
Lysine 2.16 ± 0.05 1.09 ± 0.08 2.82 ± 0.06
Methionine 0.29c ± 0.02 0.14c ± 0.01 0.65 ± 0.02
Phenylalanine 1.15 ± 0.07 0.80 ± 0.01 1.92 ± 0
Threonine 1.90 ± 0.03 1.96 ± 0.18 1.67 ± 0
Valme 2.34 ± 0.08 1.22 ± 0.17 2.26 ± 0

Total EAA 16.14 9.68 17.98

Nonessential (NEAA) (%)

Alanine 11.32 ± 0.82 2.39 ± 1.15 2.57 ± 0.02
Aspartate 3.43 ± 0.15 2.23 ± 0.27 3.74 ± 0.02
Glutamate 5.58 ± 0.64 3.21 ± 0.37 4.37 ± 0.02
Glycine 2.81 ± 0.15 1.28 ± 0.18 1.90 ± 0.02
Proline 1.62 ± 0.02 0.69 ± 0.01 2.22 ± 0.13
Serine 3.12 ± 0.13 1.76 ± 0.31 1.69 ± 0.02
Tyrosine 2.17 ± 0.15 1.52 ± 0.30 2.74 ± 0.14
Total NEAA 30.05 13.08 19.23
Total AA analyzed, % DM 46.19 22.76 37.21
Total AA as % of CPd 90.84 52.14 89.02
EAA/NEAA ratio
0.54 0.74 0.93
Crude protein (%) 50.85 43.65 41.80
a
Samples collected in 1993.
b
Adapted from original data by Phelps et al. (1975).
c
Methionine values may be underestimated as samples not subjected to prior oxidative treatment.
d
CP = crude protein.

[1984] but are within the range (25.0 - 74.0%) reported by the same authors for eight other species of
Brazilian termite soldiers and considerably lower than those reported for epigeal termite soldiers (72.0%)
from West Malaysia [Matsumoto, 1976].
The CP values are based on determination of total nitrogen (N x 6.25) and include the N present as
a component of the insects’ exoskeletons (chitin) which may not be available to the animals. Thus, the CP
values listed in Table 4 may. not represent a true digestible protein value. The protein quality of a feed is
related to the AA pattern and the availability of the AA present in the protein to the digestive process of the
animal.
All AA, dispensable and indispensable, was proportionately lower in termite soldiers than in
tamandua stomach contents, with the sole exception of threonine, which was almost identical (1.9%) (Table
4). The AA profiles of the tamandua stomach contents and M. falciger alates compared relatively well in
terms of percentage of CP and of AA analyzed. These two profiles also showed remarkable similarities in the
individual AA concentrations, except for the substantially higher level of alanine in the stomach contents
(11.3 vs. 2.6%), representing 24.5% of the total AA analyzed value. Alanine concentration was consistently
Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
Nutrition of the Tamandua 519

high in all stomach content samples, and, as termites appear to have a much lower content, its presence could
be explained as coming from ants and/or nest material, both of which were present in substantial amounts in
all tamandua stomach contents examined.
The total AA analyzed values, expressed as percentages of the DM, were much higher in the
stomach contents (46.2%) than in the termite soldiers (22.7%), likely due to protein hydrolysis and digestive
enzymes. The differences are even more striking when the values are expressed as a percentage of the CP
(90.8 vs. 52.14%). The proportion of essential and nonessential AA (EAA/NEAA ratio) indicated a much
better profile in the Nasutitermes soldiers and Macrotermes alates, with 43% and 48% of the total analyzed
value being essential AA, compared with only 35% in the stomach contents.
Unfortunately, insufficient samples prevented AA analysis of the other termite castes (workers and
alates) which were consistently present in higher proportions than soldiers in all live and roadkill stomach
contents collected and examined.
All three essential AA profiles listed in Table 4 are comparable to those recommended for dogs and
cats [National Research Council, 1985, 1986], with a possible exception of the sulphur-containing AA.
The two primary factors in defining the quality of a protein feedstuff are AA balance and AA
availability. Meat protein is illustrative of a high quality protein, as it has a protein content that is highly
digestible and an AA pattern similar to the actual AA requirements of carnivorous species. Phelps et al.
[1975] reported that the digestibility of termite protein (M. falciger alates) was poor (<50%) compared to
that of casein when fed to white rats. However, we can only speculate that the availability of these AA after
digestion in the digestive tract of a specialized anteater like the tamandua is probably much higher.
The difference between the true protein content, expressed as the total AA content, and the CP
values represents mostly nonprotein nitrogen (NPN) from the insects cuticle or exoskeleton.

Nutrient Composition of Tamandua Stomach Contents vs. Termites

Two samples of stomach contents were examined for identification purposes. The first sample
contained 69% ants, 22% termite workers, and 9% termite soldiers. The second sample consisted of 77%
termite workers, 18% soldiers, and 5% ants. This sample also contained a substantial quantity of nest
material.
The proximate analysis and GE, fiber, mineral, and vitamin concentration in stomach contents from
tamandua are compared to the overall nutrient content in termites. The results are summarized in Table 5. All
values are expressed on a DM basis.
The values presented in Table 5 indicate a remarkable similarity of the nutrient profiles between
stomach contents from Tamandua tetradactyla and the overall mean composition of Nasutitermes spp.
termites except for noticeable differences in the ADF content (31.3 vs. 25.1%) and ash content (13.9 vs.
4.1%) between the tamandua stomach contents and the termites. The lower GE value of the tamandua
stomach contents (4.6 kcal/g) was consistent with its higher ash content. The similarities in composition
between tamandua stomach contents and termites may indicate that either these comprise their main food
source or that other foods (ants) selected by tamanduas have similar compositions.

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
520 Oyarzun et at.

TABLE 5. Nutrient concentrations in tamandua stomach contents (Tamandua tetradactyla) as

compared to the overall mean composition of termites (Nasutitermes spp.)*
Stomach contents Termites overall
n Mean :t: SEM Mean :t: SEM
Drymatter(%) 10 17.77:t:1.14 29.36:t:4.32
Gross energy (kca1/g) 4 4.58 :t: 0.53 6.01 :t: 0.46
Crude protein (%) 10 50.85 :t: 1.64 58.20:t: 3.67
Crude fat (%) 9 11.20 :t 2.89 15.04 :t: 8.60
Neutral detergent fiber (%) 4 32.26 :t: 0.8 30.56 :t: 4.09
Acid detergent fiber (%) 6 31.32 :t: 2.68 25.09 :t: 4.51
Cellulose (%) 4 11.62:t: 1.13 9.77:t: 1.71
ADF-N (%) 6 3.44 :t: 0.53 3.16:t: 0.66
Lignin (%) 4 16.13 :t: 0.77 17.25 :t: 3.19
Ash (%) 9 13.85 :t: 2.72 4.11 :t: 0.23
Calcium (%) 7 0.11 :t: 0.03 0.26 :t: 0.04
Phosphorus (%) 7 0.41 :t: 0.04 0.38 :t: 0.04
Magnesium (%) 7 0.10 :t: 0.01 0.14 :t: 0.01
Potassium (%) 7 0.52 :t: 0.06 0.54 :t: 0.06
Sodium (%) 7 0.29 :t: 0.06 0.17 :t: 0.04
Iron (ppm) 7 2,748 :t: 750 652 :t: 194
Zinc (ppm) 7 190 :t: 22 163 :t: 8
Manganese (ppm) 7 82 :t: 21 57 :t: 20
Copper (ppm) 7 28 :t: 2.68 38 :t: 8
Selenium (ppm) 7 3.75 :t: 2.75 0.51 :t: 0.18
Retinol (IJ.g/g) 5 2.52 :t: 0.73 7.42 :t: 6.49
α-tocopherol (IJ.g/g) 5 44.35 :t: 11 92. 50 :t: 32.63

*All values (mean :t: SEM) are expressed on a DM basis. n = number of observations.

Termite Ca values (0.26%) were much higher than those found in the stomach contents (0.11%),
while P was equivalent at 0.4%. K was the highest among the macro minerals at 0.5%, while Fe was
predominant among the trace minerals. The concentration of this element in the stomach contents (2,748
ppm) was four times higher than in termites (652 ppm). The relatively low concentration of Ca in both
stomach contents and termites may be indicative of a lower requirement for this element in tamandua
compared to other mammals.
Retinol and α-tocopherol concentrations in termites (7.4 and 92.5 µg/g, respectively) exceeded by
more than twofold the levels in the stomach contents (2.5 and 44.4 µg/g, respectively).
The high ash (13.9%) and ADF (31.3%) and lower CP and fat concentrations in the stomach
contents most likely reflect intake of nonfood items of low digestibility.
The mineral profiles, except for Fe and Se, also show great similarity. The higher concentrations of
these two minerals in the stomach content samples may be explained either through other sources rich in
these elements (soil, nest material).

Fatty Acid Profiles of Tamandua Stomach Contents and Termite Lipids

The data on fatty acid composition of the tamandua stomach content lipids were pooled compared
to the overall fatty acid profile of the Nasutitermes spp. termites (Table 6). The values indicate that the fat
was more unsaturated in termites, with higher levels of 18:1, 18:3, and 20:4(66.2% unsaturated in termites
vs. 59.3% in the stomach contents). Fat content in termites was slightly higher (15.0%) than that in the
stomach contents (11.2%).

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
Nutrition of the Tamandua 521

TABLE 6. Fatty acid composition of tamandua (T. tetradactyla) stomach contents

and termite (Nasuitermes spp.) lipids (mean ± SEM)
Stomach contents Overall termites
Fatty Acida (%) (n = 8) (n = 6)
Capric 10:0 0.66 ± 0.10 0.55 ± 0.50
Lauric 12:0 0.51 ± 0.16 0.11 ± 0.05
Myristic 14:0 2.03 ± 0.19 4.18 ± 0.71
Myristoleic 14:1 (n-7) 0.35 ± 0.10 0.92 ± 0.73
Palmitic 16:0 23.95 ± 2.44 13.47 ± 5.69
Palmitoleic 16:1 (n-7) 2.42 ± 0.71 1.03 ± 0.39
Stcaric 18:0 11.17 ± 1.51 12.56 ± 0.12
Oleic 18:1 (n-9) 41.29 ± 4.18 47.86 ± 3.22
Linoleic 18:2 (n-6) 13.16 ± 1.83 11.90 ± 3.39
Linolenic 18:3 (n-3) 1.05 ± 0.29 2.59 ± 2.59
Arachidic 20:0 1.39 ± 0.17 1.66 ± 0.36
Gadoleic 20:1 (n-11) 0.07 ± 0.07 0.10 ± 0.10
Arachidonic 20:4 (n-6) 0.98 ± 0.32 1.77 ± 1.72
Behenic 22:0 0.49 ± 0.14 0.93 ± 0.60
SATb 40.19 ± 2.48 33.44 ± 5.12
MONOb 44.14 ± 3.71 49.91 ± 2.99
PUFAb 15.20 ± 2.08 16.25 ± 7.69
Fat soxhlet (%) 11.20 ± 2.89 15.04 ± 8.60
a
Fatty acid content expressed as percentages of total methyl esters. Fatty acids are denoted by their common name,
number of carbons:number of double bonds, followed by the position of the first double bond relative to the methyl-end
(n-).
b
SAT = sum total percentage of 10:0, 12:0, 14:0, 16:0, 18:0, 20:0 and 22:0; MONO = sum total percentage of 14:1, 16:1,
18:1, and 20:1; PUFA = sum total percentage of 18:2, 18:3, and 20:4.

Chemical Analysis of the Carton Nest Material

The chemical composition of the arboreal carton nest material of Nasutitermes spp. Is shown in
Table 7. Analysis of the two nest material samples collected at different locations indicated similar values for
DM, CP, fat, ash, macro minerals, and GE contents. However, some major differences were found in the
trace mineral and fiber fractions. The nest from Hato Masaguaral had twice as much Zn and Mn (43 vs. 24
ppm and 145 vs. 73 ppm, respectively), and a much higher Fe content (912 vs. 247 ppm) than the nest from
Hato Piñero. The lignin content in the latter was almost two times higher (71.8% vs. 47.0%). The ADF
values were higher than the neutral detergent fiber (NDF) values, for which we have no explanation. These
materials contained no measurable amounts of fat. Ca concentration was consistently high in both samples
(1.7%), but P levels were very low (0.1%).
The carton nest material of Nasutitermes termites is characterized by a total absence of fat, low
protein (7.3%), and very high lignin (59.4%). Although it has a high gross energy (5.3 kcal/g), it probably
has little or no feed value for the tamandua. This material was found in all samples of tamandua stomach
contents, and it may represent a source of Ca for this species if it is in a digestible form.

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
522 Oyarzun et al.

TABLE 7. Chemical composition of the carton nest material from Nasutitermes spp. termites*
Nest 4 Nest 5
Hato Piñero Hato Masaguaral Mean ± SEM
Drymatter(%) 80.67 81.84 81.26±0.59
Crude protein (%) 8.4 6.17 7.29 ± 1.11
Crude fat (%) 0 0 0
NDF (%) 76.35 50.74 63.55 ± 12.8
ADF (%) 81.22 80.67 80.95 ± 0.28
Cellulose (%) 2.81 2.39 2.60 ± 0.21
ADF-N (%) 1.77 1.33 1.55 ± 0.22
Lignin (%) 71.77 47.02 59.40 ± 12.4
Gross energy (kcal/g) 5.30 5.32 5.31 ± 0.01
Ash (%) 6.87 6.6 6.69 ± 0.09
Calcium (%) 1.77 1.66 1.72 ± 0.06
Phosphorus (%) 0.10 0.09 0.10 ± 0.01
Magnesium (%) 0.3 0.19 0.25 ± 0.05
Potassium (%) 0.62 0.53 0.58 ± 0.04
Sodium (%) 0.02 0.02 0.02 ± 0
Iron (ppm) 247 912 580 ± 333
Zinc (ppm) 24 43 34 ± 9.5
Manganese (ppm) 73 145 109 ± 36
Copper (ppm) 13 9 11 ± 2.2
*All values expressed on a DM basis.

Application of Findings

Since 1987, tamanduas at MTZ have principally been fed the zoo’s carnivore ration, which we
believe contains excessive quantities of Ca and vitamins for this species. Development and refinement of an
improved diet mimicking the nutrient levels identified and reported in this study and based on our own
experiences and those of other institutions maintaining tamanduas is being undertaken.

CONCLUSIONS
1. Analysis of the nutrient composition of Nasutitermes termites from Venezuela revealed them to
be a high protein, moderate fat, variable vitamin, and low mineral resource.
2. Differences in composition were seen between the various termite castes, particularly between
alates and the worker/soldier castes. Some values differed from previously reported data.
3. Stomach contents taken from T. tetradactyla showed a similar nutrient ctmposition to the overall
mean composition of termites, with some exceptions.
4. Diets consumed by free-ranging tamanduas contained 50.9% CP, 11.2% fat, 13.9% ash, 31.3%
ADF, 4.58 kcal/g GE, 0.11% Ca, 0.41% P, 2.52 µg/g retinol, and 44.35 µg/g α-tocopherol on a DM basis.
Duplication of these nutrient profiles might greatly benefit captive health and reproduction of this species.

ACKNOWLEDGMENTS

This work was funded by the Conservation Fund of the Metropolitan Toronto Zoo. It would not
have been possible without the invaluable assistance provided by many people. In particular we would like to
recognize Antonio Branger and Edgar Useche (Fundación Branger, Hato Pinero); Tomás Blohm (Hato
Masaguaral); Dr. Pedro Trebbau, Esmeralda Mujica, Marian Diaz, and Mary Cruz Lamas (Fundación

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
Nacional de Pargues Zoológicos y Acuarios de Venezuela); Dr. Carlos Bosque, Maribel Hernandez, and
Nutrition of the Tamandua 523

Maria Teresa Hosque (Universidad Simon Holivar); Ing. Agr. Mariela Romero de Lopez, Dr. José Manuel
Pernalete, and the staff of the Parque ZoolOgico Miguel Romero Antoni; Elizabeth Frank (Milwaukee
County Zoo); Sue Crissey (Hrookfield Zoo); Anne Ward (Fort Worth Zoo); José Luis Mendes Arocha,
Salvador Hoer, Efrain Siso, and Juan Sierra (Profauna, Ministerio del Ambiente y Recursos Naturales
Renovables de Venezuela); and Mr. Jesus Coty (Universidad de Maracaibo).

REFERENCES

AOAC. OFFICIAL METHODS OF ANALYSIS OF THE ASSOCIATION OF OFFICIAL ANALYTICAL

CHEMISTS, 12th ed. Washington, DC, Association of Official Analytical Chemists, 1975.

AOAC. OFFICIAL METHODS OF ANALYSIS OF THE ASSOCIATION OF OFFICIAL ANALYTICAL

CHEMISTS, 13th ed. Washington, DC, Association of Official Analytical Chemists, 1980.

AOAC. OFFICIAL METHODS OF ANALYSIS OF THE ASSOCIATION OF OFFICIAL ANALYTICAL

CHEMISTS, 15th ed. Arlington, VA, Association of Official Analytical Chemists, 1990.

Crawshaw, G.J.; Oyarzun, SE. Vertebral hyperostosis in anteaters: Probable hypervitaminosis A and/or D.
JOURNAL OF ZOO AND WILDLIFE MEDICINE, 27:158—169, 1996.

Dierenfeld, E.S.; Barker, D.; McNamara, T.S.; Walberg, J.A.; Furr, H.C. Vitamin A and insectivore
nutrition. VERH. BER. ERKRG. ZOOTIERE 37:245—249, 1995.

Emerson, R.J. A modification in the official methods for the determination of metals in feeds and fertilizers
by atomic absorption spectrophotometry. JOURNAL OF THE ASSOCIATION OF OFFICIAL
ANALYTICAL CHEMISTS 58:158, 1975.

Frye, FL.; Calvert, C.C. Preliminary information on the nutritional content of mulberry silk moth (Bombyx
mori) larvae. JOURNAL OF ZOO AND WILDLIFE MEDICINE 20:73-75, 1989.

Hoffman, J.; Westerly, R.J.; Hidiroglou, M. Precise fluorimetric microdetermination of selenium in

agricultural materials. JOURNAL OF THE ASSOCIATION OF OFFICIAL ANALYTICAL CHEMISTS
51:1039—1042, 1968.

Lubin, Y.D.; Montgomery, G.G. Defenses of Nasutitermes termites (Isoptera, Termitidae) against tamandua
anteaters (Edentata, Myrmecophagidae). BIOTROPICA 13:66—76, 1981.

Lubin, Y.D.; Montgomery, G.G.; Young, OP. Food resources of anteaters (Edentata:Myrmecophagidae). I.
A year’s census of arboreal nests of ants and termites on Barro Colorado Island, Panama Canal Zone.
BIOTROPICA 9:26—34, 1977.

Matsumoto, T. The role of termites in an equatorial rain forest ecosystem of West Malaysia. I. Population
density biomass carbon nitrogen and calorific content and respiration rate. OECOLOGIA 22:153 - 178,
1976.

Meritt, D.A. Edentate diets currently in use at Lincoln Park Zoo, Chicago. INTERNATIONAL ZOO
YEARBOOK 10:136—138, 1970.

Meritt, D.A. The lesser anteater (Tamandua tetradactyla) in captivity. INTERNATIONAL ZOO
Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.
YEARBOOK 15:41—45, 1975.

Menu, D.A. The nutrition of edentates. INTERNATIONAL ZOO YEARBOOK 16:38-46, 1976.

Montgomery, G.G. Impact of vermilinguas (Cydopes, Tamandua:Xenarthra = Edentata) on arboreal ant

populations. Pp. 351—363 in THE EVOLUTION AND ECOLOGY OF ARMADILLOS, SLOTHS, AND
VERMLLINGUAS. G.G. Montgomery, ed. Washington, DC, Smithsonian Institution Press, l985a.

Montgomery, G.G. Movements, foraging and good habits of the four extant species of neotropical
Vermilinguas (Mammalia: Myrmecophagidae). Pp. 365—377 in THE EVOLUTION AND ECOLOGY OF
ARMADILLOS, SLOTHS, AND VERMILINGUAS. G.G. Montgomery, ed. Washington, DC, Smithsonian
Institution Press, 1985b.

Montgomery, G.G.; Lubin, Y.D. Prey influences on movements of neotropical anteaters. Pp. 103 – 131 in
PROCEEDINGS OF THE 1975 PREDATOR SYMPOSIUM. R.L. Phillips; C. Jonkel, eds. Missoula, MT,
University of Montana Printing Department, 1977.

National Research Council. NUTRIENT REQUIREMENTS OF DOGS. Washington, DC, National

Academy Press, 1985.

National Research Council. NUTRIENT REQUIREMENTS OF CATS. Washington, DC, National

Academy Press, 1986.

Phelps, R.J.; Struthers, J.K.; Moyo, S.J.L. Investigations into the nutritive value of Macrotermes falciger
(Isoptera:Termitidae). ZOOLOGICA AFRICANA 10:123—132, 1975.

Redford, K.H. Feeding and food preference in captive and wild giant anteaters (Mymecophaga tridactyla).
JOURNAL OF ZOOLOGY 205: 559 - 572, 1985.

Redford, K.H.; Dorea, J.G. The nutritional value of invertebrates with emphasis on ants and termites as food
for mammals. JOURNAL OF ZOOLOGY 203:385—393, 1984.

Sarwar, G.; Botting, H.G.; Peace, R.W. Complete amino acid analysis in hydrolisates of foods and feces by
liquid chromatogrpahy of precolumn phenylisothiocyanate derivatives. JOURNAL OF THE
ASSOCIATION OF OFFICIAL ANALYTICAL CHEMISTS 71:1172, 1988.

Van Soest, P.J. NUTRITIONAL ECOLOGY OF THE RUMINANT. Corvallis, OR, 0 & B Books, 1982.

Van Soest, P.J. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation
to animal nutrition. JOURNAL OF DAIRY SCIENCE 74:3583 - 3597, 1991.

Wetzel, R.M. The species of tamandua Gray (Edentata, Myrmecophagidae). PROCEEDINGS OF THE
BIOLOGICAL SOCIETY OF WASHINGTON 88:95—112, 1975.

Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents
From Wild Tamanduas (Tamandua tetradactyla). S.E. Oyarzun, G.J. Crawshaw, E.V. Valdes. Zoo Biology.
Copyright©1996 Wiley-Liss, Inc. Reproduced with permission of John Wiley & Sons, Inc.